FORD GO BIKE DATA ANALYSIS

by Aswathy Nair

Investigation Overview

In Ford Go Bike Data Analysis,I wantto investigate factors affecting Renting bikes. The variables considered are Monthly,Weekdays, Hours, User Types, Gender, Duration and Age groups that Bikes.

Dataset Overview

GoBike System Data is used for analysis and visualization. This data set includes information about individual rides made in a bike-sharing system covering the greater San Francisco Bay area.

Note that the above cells have been set as "Skip"-type slides. That means that when the notebook is rendered as http slides, those cells won't show up.

Trip Duration Distribuition

Distribution is skewed right and clearly shows most of the trips have duration of 300 - 700 seconds.

bins_edge = np.arange(0, df.duration_sec.max()+100, 100)

plt.figure(figsize=(12,8))
plt.hist(data=df, x='duration_sec', bins=bins_edge)
plt.xlim(0, 2000)
plt.xlabel('Duration (sec)',fontsize = 14);
plt.ylabel('Frequency', fontsize = 14);
plt.title('Trip duration distribution',fontsize=14);

Distribution of Age Group on Rides

Distribution shows major subscribers belong to age group 30 - 40

fig = plt.figure()
base_color = sb.color_palette()[0]
plt.figure(figsize=(12,8))
_ = sb.countplot(data = df, x = 'age_group', color = base_color)
plt.xlabel('Age Group',fontsize = 14);
plt.ylabel('Frequency',fontsize = 14);
plt.title('Age Group on Bike Ride',fontsize = 14);

<Figure size 432x288 with 0 Axes>

Bike Usage vs Hour of Day

Distribution indicates 8- 9 AM and 4-6 PM is the peak hours for the ride.

fig = plt.figure();
plt.figure(figsize=(14,8))
base_color = sb.color_palette()[0]
_ = sb.countplot(data=df, x='start_hour',palette='viridis')
plt.xlabel('Trip start hour',fontsize=14);
plt.ylabel('Total rides',fontsize =14);
plt.title('Ford GoBike usage by Hour of day',fontsize = 14);

<Figure size 432x288 with 0 Axes>

Distribution of Bike rides for Gender and User types

Male gender are the main users of bike rides and of them major group belongs to subscibers

fig = plt.figure();
plt.figure(figsize=(14,8))
sb.countplot(data=df, x='member_gender', hue='user_type', palette='viridis')
plt.xlabel('Gender Type',fontsize = 14);
plt.ylabel('Frequency',fontsize = 14);
plt.title('Frequency vs Gender for User Type',fontsize = 14);

<Figure size 432x288 with 0 Axes>

Distribution of Ride Duration Monthly

July 2017 has highest duration for rides considering Ford Go Bikes was launched in June 2017. Duration got and have monthly average of 750 seconds

plt.figure(figsize = [20, 10]);
base_color = sb.color_palette()[1]
sb.pointplot(x="year_month", y="duration_sec",data=df.sort_values(by='year_month'), color=base_color);
plt.xticks(rotation = 90);
plt.xlabel("Year-Month",fontsize = 14);
plt.ylabel("Duration (sec)",fontsize = 14);
plt.title("Year-Month vs Duration (sec) distribution",fontsize = 14);

Distribution of ride duration for Age groups on Weekdays

Duration of rides are more during weekends especially for older age group

plt.figure(figsize = [30, 26]);
day_order = ['Mon','Tue','Wed','Thu','Fri','Sat','Sun']
g = sb.FacetGrid(data = df, col = 'age_group',col_wrap = 2,height = 5, aspect=1.5,margin_titles=True);
g.map(sb.boxplot, 'trip_weekday', 'duration_sec',order = day_order, color="c")

g.fig.subplots_adjust(wspace=.2, hspace=.2);
g.fig.subplots_adjust(top=.9);
g.set_titles(col_template = "Age Group - {col_name}",fontsize = 14);
g.fig.suptitle('Age Group Wise Distribution of Duration Across Weekdays ',fontsize = 14);
g.set_xlabels('Day of Week',fontsize = 14);
g.set_ylabels('Duration (sec)',fontsize = 14);

plt.ylim(0,3000);
plt.show();

<Figure size 2160x1872 with 0 Axes>

Distibution on usage duration on Gender and User type on Monthly

Customers ride duration is higher than subscribers and female gender ride more than male gender.

plt.figure(figsize=[20, 8])

plt.subplot(1, 2, 1)
sb.pointplot(data=df.sort_values(by='year_month'), x='year_month', y='duration_sec', hue='user_type', palette= 'viridis')
plt.xticks(rotation=90);
plt.ylim(0, 3000);
plt.xlabel('Year Month',fontsize = 14);
plt.title('Usertype vs Duration (sec) for Year month');
plt.ylabel('Duration',fontsize = 14);

plt.subplot(1, 2, 2)
sb.pointplot(data=df.sort_values(by='year_month'), x='year_month', y='duration_sec', hue='member_gender', palette='viridis')
plt.xticks(rotation=90);
plt.ylim(0, 3000);
plt.xlabel('Year Month',fontsize = 14);
plt.ylabel('Duration',fontsize = 14);
plt.title('Gender vs Duration (sec) for Year month');